Project 4: Optical monitoring of post-translational modification of proteins in vivo (Thorn)(#27) We developed tools to study protein modification and interactions using fluorescence resonance energy transfer (FRET) microscopy. The main effort was to detect protein-protein interactions in wVo12 and use these data to predict the structure of large protein complexes. We developed methods to produce precise and reproducible FRET data from yeast, and used them to derive a low-resolution structure for the septin complex (which is required for cell division from yeast to mammals). With Michael Laub (project 7) we found interactions between two-component signaling proteins, the predominant participants in post-translational regulation in bacteria, and collaborated with Daugherty and his colleagues to develop high throughput, quantitative, FRETbased methods to identify protein interactions13. Kurt Thorn, the PI, was a Bauer Fellow and is now Director of the Nikon Imaging Center at UCSF. New project 4: Tandem repeats in genes: hyper-variable modules for adaptation? (Verstrepen) (#28-30) We tested the hypothesis35 that in budding yeast, tandemly repeated sequences are hyper-variable genetic modules, whose variability helps organisms adapt to changing environments. Most sequenced genomes contain tandem repeats;some of these occur in coding sequences, where recombination can alter the number of peptide repeats in translated proteins. We have, 1) developed an algorithm36 (http://hulsweb1.cgr.harvard.edu/SERV/) to find and annotate repeats, predict whether they vary in number, associate them with other information (expression level, overlap with transcription factor binding sites, nucleosome sites etc.), and deposit this data in a publicly accessible database (http://hulsweb1.cgr.harvard.edu/TandemRepeat/) that allows rapid and automated comparison to existing databases;2) generated yeast strains with different numbers of repeats in the promoters and open reading frames of selected genes, and studied the phenotypic consequences of this variation (ongoing);3) studied tandem repeats in Arabidopsis thaliana and Oryza sativa (with C. Queitsch, new project 7). We have used two sequenced genomes to find all ORFs containing variable numbers of repeats, selected 150 genes, characterized their variation across twelve A. thaliana isolates from different locations, and found more than 70 genes that show variation in repeat number between different isolates. Kevin Verstrepen became a Bauer Fellow in 2005 and joined the CMB in 2006.